Adaptive Best-of-Both-Worlds Algorithm for Heavy-Tailed Multi-Armed Bandits.

Jiatai Huang Yan Dai Longbo Huang

Published in: CoRR (2022)

Keyphrases

computational complexity
probabilistic model
k means
np hard
expectation maximization
learning algorithm
dynamic programming
similarity measure
optimal solution
lower bound
multiresolution
active learning
markov random field
generalized gaussian
multi armed bandits